WO2011022884A1

WO2011022884A1 - Beam-forming based uplink system resource allocation method and device

Info

Publication number: WO2011022884A1
Application number: PCT/CN2009/073536
Authority: WO
Inventors: 谢玉堂; 赵新胜; 方伟
Original assignee: 中兴通讯股份有限公司
Priority date: 2009-08-26
Filing date: 2009-08-26
Publication date: 2011-03-03

Abstract

A beam-forming based uplink system resource allocation method and device are provided, the method includes following steps: A. a base station (BS) obtains channel state information in a frame of each uplink time slot of each subscriber station (SS), and combines the decomposed maximum eigenvalues with the corresponding eigenvectors to form a channel power gain matrix of each time slot of each SS; B. the BS allocates the time slot to each SS; and the BS sends the allocated time slot and the beam-forming vector corresponding to the time slot to each SS; C. each SS extracts the allocated time slot information, and allocates power and bits; D. each SS sends the allocated power and bit information to the BS; the BS processes received data, and obtains original data sent from the SS; in each allocation period of subcarrier, power and bits allocation, step A to step D are repeated. By using the invention, transmission power of the base station is minimized, while transmission power minimization of each SS is ensured, and system performance is improved.

Description

The present invention relates to the field of wireless communications, and in particular, to a beamforming based uplink system resource allocation method and apparatus. BACKGROUND OF THE INVENTION Current research on adaptive allocation algorithms mainly focuses on the downlink aspect, and there are few researches on uplink adaptive allocation algorithms, and mainly for uplink adaptive allocation of Orthogonal Frequency Division Multiplexing (OFDM) systems. Algorithm research, because the maximum difference between uplink and downlink is distributed power limitation, that is, each user has its own power limitation in uplink, so the research on uplink adaptive allocation algorithm has certain difficulty compared to downlink, how to target uplink An effective adaptive allocation scheme is proposed to make better use of channel resources and meet the needs of different services, which is of great significance for the next generation of mobile communication systems. In the patent application with the application number 200410068029.X, Shanghai Jiaotong University proposed a power minimization allocation method combining water injection algorithm and greedy algorithm. The method firstly uses the water injection algorithm to perform initial power allocation for each subchannel, and then uses the greedy algorithm. The remaining power is redistributed. Although this method can make the allocation result better, the complexity is higher, and it is allocated for the single client. In the patent application with the application number 200510000082.0, Beijing University of Posts and Telecommunications proposes a power minimization allocation method for an OFDM system, which first divides all subchannels into equal groups at intervals, and then in each group, simultaneously Perform power and bit allocation. Although the method is very low in complexity, the method is for single-user allocation, not for practical systems, and does not consider multiple antennas. In the patent application with the application number 200510083831.0, Beijing University of Posts and Telecommunications adopts a low-complexity power minimization allocation method, which includes adaptive modulation and adaptive demodulation. For adaptive modulation, only in the transmission The power and bit allocation are performed for one frame, and then the allocation scheme is saved. When the subsequent frame is sent, the stored scheme is called for corresponding mapping. Although the scheme is simple, the real-time performance is not good, especially in the channel condition. When the change is fast, such as high-speed moving conditions, the performance is poor. SUMMARY OF THE INVENTION In view of this, the main object of the present invention is to provide a method for allocating an uplink system resource based on beamforming, which is applicable to resource allocation of a single-cell multi-user uplink adaptive system, which has low complexity and good real-time performance. To achieve the above objective, the technical solution of the present invention is implemented as follows: According to an aspect of the present invention, a method for allocating an uplink system resource based on beamforming is first provided, the method comprising:

A. The base station BS obtains channel state information of each uplink of each user SS in one frame by channel sounding technology, and performs eigenvalue decomposition on the multiple input multiple output MIMO channel matrix for each SS in each time slot, and obtains corresponding values. The eigenvector and the eigenvalue, the maximum eigenvalue obtained after the decomposition and the corresponding eigenvector form a channel power gain matrix of each slot of each SS;

B. The BS allocates time slots to each SS according to the channel power gain matrix, and uses the marginal utility and the improved greedy algorithm; the BS sends the allocated time slots and the beamforming vectors corresponding to the time slots to the SSs;

C. Each SS extracts the allocated time slot information from the received downlink subframe, and performs power and bit allocation;

D. Each SS sends the allocated power and bit information to the BS through the control part in the uplink subframe; the BS processes the received data to obtain the original data sent by the SS; in each allocation period of the subcarrier, power, and bit allocation Repeat steps A through D. Step A specifically includes: Al, the SS sends a service flow parameter to the BS when the service is customized;

A2. In each allocation period, the BS acquires channel state information of each time slot between the BS and each SS by using channel sounding technology, and then performs eigenvalue decomposition on the MIMO channel matrix of each time slot of each SS to obtain a corresponding The feature vector and the eigenvalue are selected, and the largest eigenvalue and the corresponding eigenvector are selected to form a channel power gain matrix H of each SS slot, a transmitter beamforming vector corresponding to the matrix, and a merging vector. The above time slot means: Using the adaptive modulation and coding AMC subcarrier arrangement in the uplink, each time slot is composed of 6 block bins and 1 orthogonal frequency division multiple access access OFDMA symbol, and 1 bin is composed of 9 sub-frames of the OFDMA symbol. The carrier component, wherein 8 subcarriers are data subcarriers, and the other one is a pilot subcarrier. The above step B specifically includes:

Bl. When each spatial subchannel on each time slot is transmitted in a maximum modulation order, the minimum number of slots required for each SS is calculated;

B2, first allocate a time slot for each SS;

B3. Find 4 out-of-order time slots in each SS unassigned time slot, and calculate the marginal utility of the time slot;

B4. Allocating the required number of slots for each SS under the condition that each SS can obtain the minimum number of slots required;

B5. After determining step B4, whether there is a time slot remaining, if any, allocate the remaining time slots to the corresponding SS; otherwise, perform step C. The above step C specifically includes:

C1, SS uses the improved greedy algorithm for power and bit allocation on the allocated time slots;

C2, SS uses the allocation result in C1 to perform corresponding modulation and power adjustment, and weights the modulated symbols according to the corresponding beamforming vector, and then undergoes inverse fast Fourier transform (IFFT), parallel-to-serial conversion and adding force. The prefix is sent to the BS. The above step D specifically includes:

Dl. The receiving end first performs serial-to-parallel transformation and removes the cyclic prefix on the received data, and then performs fast Fourier transform FFT transform into frequency domain symbols, and the receiving end performs combined vector processing on the transformed frequency domain symbols;

D2, demodulating the original data by using the received power and bit allocation information and the time slot allocation information. According to an aspect of the present invention, an uplink system resource division based on beamforming is also provided. With the device. A beamforming-based uplink system resource allocation apparatus according to the present invention, the apparatus comprising an adaptive beamforming and subcarrier allocation module and a power and bit allocation module, wherein the adaptive beamforming and subcarrier allocation module is based on channel state information, The eigenvalue decomposition of the MIMO channel matrix is performed on each time slot of each SS to obtain a corresponding feature vector and eigenvalue, and the largest eigenvalue and the corresponding eigenvector are selected to form a channel power gain matrix H and a matrix H of each SS slot. Corresponding transmitter beamforming vector and receiver combining vector; transmitting beamforming vector and subcarrier allocation information to the SS side through the BS control channel and the BS antenna; the power and bit allocation module according to the beamforming vector and subcarrier allocation transmitted by the BS Information, using a marginal utility and an improved greedy algorithm to assign time slots to each SS. The present invention is directed to an adaptive allocation requirement of an uplink multiple input multiple output (MIMO) OFDM system, and proposes a single-cell multi-user uplink adaptive subcarrier, power and bit allocation method. The present invention separates the subcarrier allocation from the power and the bit allocation under the condition of ensuring the monthly quality requirements and fairness of each client (SS, Subscriber Station), so that the base station (BS, Base Station) transmits the minimum power, and at the same time Ensure that each SS has the lowest transmit power and improve system performance. The drawings are intended to provide a further understanding of the invention, and are intended to be a part of the description of the invention. 1 is a schematic structural view of an application system of the present invention; FIG. 2 is a flow chart of the method of the present invention; and FIG. 3 is a schematic structural view of the device of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The embodiments of the present invention have an obvious effect on the improvement of system performance through adaptive allocation techniques, and the techniques of subcarrier allocation and power and bit allocation in the adaptive allocation technique are also Interrelated and mutually influential, if the combined subcarriers, power and bit allocation complexity are high, the present invention adopts a two-step method, that is, subcarrier allocation and power and bit allocation are performed separately, and the method is minimized. The transmission power of the system is targeted, and the requirements of QoS (Quality of Service), fairness between users, and dynamic changes of channel state information in actual applications are considered. In order to make the objects, technical solutions and advantages of the present invention more clear, the present invention will be further analyzed in detail below. First, the eigenvalue decomposition based beamforming principle of the MIMO-OFDM system, the improved greedy algorithm, and the representation of related symbols will be described in order to describe the present invention. Assume

The number of transmitting antennas of the MIMO-OFDM system is Wt, the number of receiving antennas is Wr, and the number of user terminals is K. The number of subcarriers is assumed to be allocated in the "exclusive" mode in the embodiment, that is, each subcarrier is allocated to only one user. end. For a MIMO-OFDM system, the receiving symbol on the receiving end, ie, the "subcarrier" of the UE, can be expressed as:

Where is the modulation symbol on the "subcarrier", ^C " ^{= [C} «> ¹ , ^C "' ² , · · · , ^C «' ^{Nt ]T} ' ^d « ^{= [d} «> ¹ , 2 d _{n> Nr} f represents the transmitter beamforming vector of Nt X 1 and the receiver combining vector of ^{χ 1} on the first subcarrier, respectively, ^Η « is a matrix, and the elements in the matrix are the nth subcarrier of each client k The frequency response of the channel between the different transmit and receive antennas, ^N " is the complex Gaussian noise vector of ^{χ 1} , and each element of ^Ν "is the mean

0. Complex Gaussian random variable with variance. At the receiving end, the way to maximize the signal-to-noise ratio is Maximum Ratio Combining (MRC). For (1), the corresponding maximum ratio combining vector is

The signal ^Z « obtained by performing maximum ratio combining on the received signal of equation (1) can be expressed as:

When the beamforming vector is selected as ^e , the signal-to-noise ratio at the receiving end can be maximized. This In it, the eigenvalue vector corresponding to the maximum eigenvalue ^Λ obtained after eigenvalue decomposition (EVD, Eigen Value Decomposition), ^Ά "the eigenvalue decomposition result is as follows:

HfH = U A Uf

(4) where ^Λ "is a diagonal matrix, ^Λ "the elements on the diagonal are the eigenvalues of the matrix ¹¹ ",

"It is a unitary matrix, ie ¹ ^ ^U « ^=I , the column of ^U « consists of the eigenvectors of the matrix ¹ ^ ^H ". Therefore, the corresponding subcarrier channel gain matrix H on the N subcarriers of the K subscribers is:

H =

(5) where, the transmit beamforming vector ^e " corresponding to each feature value of H and the reception maximum ratio combining vector d

"respectively

c„ = u

(6) d„ = (H u

(7) Set to a certain client, order ¹¹

And adopt an improved greedy algorithm to perform power and bit allocation on the client ^k . The so-called greedy algorithm means that in each bit allocation process, the subcarriers requiring the minimum incremental power are selected, and only one bit is allocated to the subcarriers at a time, and the power required for the subcarriers is calculated, and the process is repeated until all the signals are to be transmitted. The number of bits is allocated. In the improved greedy algorithm, the condition that the number of bits allocated on each subcarrier ⁶ ' is a non-negative integer is ignored, and the minimum value of the total transmitted power is calculated by the Lagrangian multiplier method:

Among them, "The number of bits that need to be transmitted by the client, and then rounded up by the following formula:

( 9 )

£H ∑b _t <R _tliIget Improved greedy algorithm is: If ' , the current flow is ended; if ' = i , then for each subcarrier l ≤ i ≤ N, calculate the addition of a bit on the subcarrier Power increase:

Δ _Α =Γσ ² A , loop through the following steps until 'to: select subcarriers

The number of bits allocated for this subcarrier is updated as: b _b „ aL , if

∑b _t >R _t:

Then for each subcarrier ί' l ≤ i ≤ N, calculate a bit band reduction on the subcarrier to perform the following steps, until: the number of bits allocated by the subcarrier is updated to:

After the above steps, the number of bits allocated on each subcarrier is ⁶ '·, and then the power allocated on each subcarrier can be obtained according to the relationship between power and bit. The relationship between power and bit is related to the bit error rate of the subcarriers. When using uncoded quadrature amplitude (QAM, Quadrature Amplitude)

Modulation) When modulating, the bit error rate of the first subcarrier is approximated as:

Where = P, ^h J. - for the signal-to-noise ratio (SNR) of the i-th subcarrier, the required transmit power of subcarrier ^ζ ' can be approximated as:

Γ is the signal-to-noise ratio difference (SNR gap), which characterizes the actual channel capacity and fragrant

The difference between the capacity of the agricultural channel, Bei 'J:

The symbols used in the present invention are described below: "Number of bits allocated for the user side on the subcarrier", ^A , "Representing the subcarrier" is assigned to the UE, ^k , " ^{= 1} indicates that the subcarrier n is assigned to The user terminal k, ^Pk ' ^{n =} ° , indicates that the subcarrier n is not allocated to the client k; the client set is ^: ² ,..., ^^ , and the subcarrier set is

S = {12, ... , N} ^ The set of subcarriers allocated by the client is recorded as the number of subcarriers contained in ^S k is the maximum modulation order supported by the system, indicating that the user side The marginal utility of the subcarriers. The present invention employs a system as shown in FIG. 1 for resource allocation, the system comprising a BS and at least one SS, wherein, on the BS side, the channel state information acquisition module estimates according to a Sounding Channel technique Channel state information of k user terminal carriers; the adaptive beamforming subcarrier allocation module performs eigenvalue decomposition on the MIMO channel matrix on each time slot of each SS according to channel state information, and obtains corresponding feature vectors and eigenvalues, and selects The largest eigenvalue and the corresponding eigenvector form the channel power gain matrix H of each slot of each UE, the transmitter beamforming vector corresponding to the matrix H, and the merging vector of the receiving end; the beamforming vector and the sub-band are transmitted through the BS control channel and the BS antenna. The carrier allocation result is sent to the SS side; on the SS side, through the SS antenna, the SS control channel, the power and bit allocation module according to the beamforming sent by the BS Subcarrier allocation amounts and results, and improved use of marginal utility for each SS greedy algorithm to allocate power and number of bits; SS transmits the service flow parameter module parameters to the BS customizing service; Then, the UE power and bit loading module of each SS allocates the bits on each subcarrier according to the maximum eigenvalue 4, l ≤ i ≤ K of the beamforming vector in the respective application layer buffer, and passes the adaptive modulator. According to the power and bit allocation result of the UE, the number of bits allocated on each subcarrier is dynamically adjusted, and converted into a frequency domain signal, and each of the SS beamforming vectors is processed by each adaptive beamforming vector module, and then subjected to fast Fourier. The inverse inverse Fourier Transform (IFFT) module is transformed into the time domain, and the prefix is added by the parallel-to-serial conversion module and the added prefix module, and the data of each SS is sent by the antenna to the BS;

On the BS side, the data of each SS is received via an antenna, the prefix is removed by a serial-to-parallel transform and a de-prefix module, and the received data is transformed into the frequency domain by a Fast Fourier Transform (FFT) module, and the vector is combined by the receiving end. The module combines the data on each subcarrier and demodulates it into a digital signal by an adaptive demodulator, and the extraction module extracts the bit information of the user end for subsequent processing by the system. In the above system, the adaptive beamforming and subcarrier allocation module on the BS side and the power and bit allocation module on the SS side are new modules of the present invention, and the allocation of subcarriers can be completed by the BS, and the power and bit allocation are performed by The SS is completed, and adaptive adjustments are made in real time to improve the efficiency of the system. The embodiments of the present invention are further described below in conjunction with the specific embodiments and the accompanying drawings: Step 1. The BS performs system initialization. Each client sends a Service Flow Parameter to the base station when the service is customized, including a transmission rate and a bit error rate required for ensuring QoS. In each allocation period, the BS estimates the channel by using the channel detection technology. Channel state information (CSI, Channel State Information), and then EVD the MIMO channel matrix on each time slot of each SS to obtain corresponding feature vectors and eigenvalues, and select the largest eigenvalue and corresponding The feature vector constitutes a channel power gain matrix H of each slot of each client, a transmitter beamforming vector corresponding to the matrix H, and a receiver merge vector. The time slot refers to: adopting the adaptive modulation and coding (AMC) subcarrier 4 in the uplink, and each time slot is composed of 6 blocks (bin) and 1 orthogonal frequency division. An OFDMA (Orthogonal Frequency Division Multiple Access) symbol is constructed, wherein one bin is composed of 9 subcarriers in an OFDMA symbol, wherein 8 subcarriers are data subcarriers, and the other subcarrier is a pilot subcarrier. Step 2: The BS allocates a time slot to each SS by using a marginal utility and an improved greedy algorithm according to the channel power gain matrix; the BS passes the allocated time slot and the beamforming vector corresponding to the time slot through the downlink subframe. The control part is sent to each SS.

The so-called marginal utility refers to the amount of power reduction that can be brought when a certain subcarrier is allocated to the user. Step 2 specifically includes: Step 201: Each spatial subchannel on each time slot is transmitted in a maximum modulation manner, and the minimum number of time slots required by each user terminal is:

Where "^ means rounding up, that is, the upper limit (ceil) function is used, and there is W≥N'

N'=∑N _t

Step 202: Allocating a time slot for each user end, and performing the following steps for each user terminal ^ ^{1 ≤ ≤} ^>: Step 202a, performing time slots of each user end in descending order of channel power gain ; n =arg<min(p,„)

Step 202b: Select the time slot with the highest ranking: ^L - ^s ' ^J , update:

Ρ ⁼¹ , }, C _k =C _k +l, S = S-{n }_, that is, all the bits to be allocated by the UE are all placed on the first allocated subcarrier, that is, "the corresponding subcarrier, Thus, the number of subcarriers allocated by the user k and the number of subcarriers added & contained are also increased by 1, and the number of unallocated subcarriers in the set of subcarriers is reduced. Step 203: Calculating the top of each user terminal 4 The marginal utility of the time slot, for each client (1 ≤ ≤, respectively, perform the following steps: Step 203a: Find the top ranked time slot among the unallocated subcarriers:

Step 203b: Calculating the marginal utility of the time slot, the calculation method of the marginal utility ^: Letting a certain user terminal have allocated β time slots, and the channel power gain on the qth time slot represents ^ ^h _q ^≤ q ^≤ Q , The number of bits allocated for each time slot is expressed as b _q ( ^l ^≤ i ^≤ Q); the newly added time slot is denoted as β + ι, and the channel power gain of time slot δ ⁺¹ is expressed as ^h Q"

Then, the calculation procedure of the marginal utility of the time slot δ + ¹ is as follows: Step 203b1, initialize ⁰ e ^{+1 =0} , ΔΡ ^{= 0} For any time slot, calculate the power that can be reduced by reducing one bit: Δρ _? =Γσ ² 23⁄4 ^l /h _q ^ where 1 _≤ 3⁄4 τ _≤ ρ; Calculate the power required to transmit one bit on time slot β + l: pQ

.,

Step 203b2: Find the time slot with the largest power reduction:

Step 203b3, if ^≤ 2 ⁺¹ , the calculation ends; otherwise, one bit of the slot ^ is transferred to the slot 2 + 1 and the following parameters are updated: ^b ₉ H i Q ^+l , i I 1 ,

^υ / e ⁺¹ , megabyte to step 203b2 After the marginal utility of the time slot β + l is calculated, 3⁄4 ( ^{1 ≤ ≤} 2 ^{+ 1} ) is the bit allocation result,

0 inch · gap β + ι the marginal utility of the client. Step 204: Under the condition that each SS can obtain the minimum number of slots required, allocate the required number of slots for each SS, and perform the following steps cyclically until 0, and =0: Step 204a, find the client with the largest marginal utility:

The corresponding time slot is "step 204b, if the number of subcarriers allocated by the user k has exceeded the minimum number of subcarriers required by the UE according to Equation 12, then the UE is not required to be provided. Continue to allocate subcarriers to save system resources, ie ^ ^{= [/} Go to step 204d; otherwise, assign the time slot to the client, ie ⁼¹ , update: }, ^C * ⁼ +^S^S- {n } , and redistribute the bits of the UE on the allocated time slots, that is, update ^"^ step 204c, if the time slot" simultaneously serves as the most advanced time slot of the plurality of client terminals 4, then It is necessary to additionally allocate time slots to the UEs assigned to the time slots. For each client that is not assigned to slot ⁿ : select the most advanced slot in the unallocated slot, and calculate the marginal utility of the slot, ie

3⁄4 =arg{min( _ft )

The top unassigned time slot of the end sort: , calculate the marginal utility of the time slot

Step 204d: Determine whether ^ and S are empty at the same time. If both ^ and S are empty, indicating that the time slot allocation is completed, step 3 is performed; otherwise, step 204a is performed. Step 205: Allocate the remaining time slots to the corresponding UE to minimize the power. The following steps are performed cyclically until S = 0, which is basically the same as the process of step 204, but here is for all SSs. First, it is determined whether S is empty after step 204. If it is empty, indicating that there are no remaining time slots, step 3 is performed; otherwise, the following steps are performed: k* = arg { max( p , )

Step 205a, find the largest marginal utility UE: ¹ J, the barrel has slots corresponding to "step 205b, the slots" assigned to the UE, i.e., ^μ "'~, Update: ^{^S} _k' ^ ^S k' ^{+ {n} }^C _k , ^C _k , +\,S^S-{n } ^ and redistribute the bits of the client on the allocated time slots, ie update," step 205c, if The time slot "simultaneously acts as the most advanced time slot of multiple subscribers 4, then additional time slots need to be allocated to the UEs allocated to the time slots. For each client that is not assigned to the time slot: Unallocated Select the 4 unordered highest time slot in the time slot, and calculate the marginal utility of the time slot,

Step 205d: Determine whether S is empty. If it is empty, perform step 3; otherwise, perform step 205a. Step 3 SS performs power and bit allocation according to the time slot allocation result sent by the BS. Step 301: Change W in equation (9) to ^max as the number of bits to be transmitted by the SS. ^ is the number of bits in the system's maximum modulation mode, that is, the maximum number of bits that the SS subcarrier can transmit. Calculate the time slots assigned to each SS using the improved greedy algorithm described earlier

Ν

∑bi >R _tt

Power value and number of bits, but here some improvement, if ' = i, then first judge whether it is greater than 0, if greater than 0, the remaining steps are the same as the improved greedy algorithm; if less than 0, then the time slot The power increment is set to infinity, and the time slot is no longer allocated bits in the current allocation period. Step 302: According to the bit number allocation result, perform corresponding modulation and power adjustment, and after modulation The symbols are weighted according to the corresponding beamforming vector. The so-called weighting refers to adjusting the antenna direction of the SS according to the angle and amplitude of the beamforming vector to obtain the best signal; then passing the IFFT, parallel transform and adding force p A series of processing such as prefixes are sent to the BS. Step 4: Each SS sends the allocated power and bit information to the BS through the control part in the uplink subframe; the BS processes the received data to obtain the original data sent by the SS. The receiving end, that is, the BS first performs serial-to-parallel conversion and removes the cyclic prefix on the received data, and then performs FFT as a frequency domain symbol, and then performs combined vector processing on the obtained frequency domain symbols; and then, uses the received power and bit allocation information. And the original time slot allocation information demodulates the original data. The present invention also provides an uplink system resource allocation apparatus based on beamforming. As shown in FIG. 3, the apparatus includes an adaptive beamforming and subcarrier allocation module and a power and bit allocation module, which are adaptive beamforming and The subcarrier allocation module performs eigenvalue decomposition on the MIMO channel matrix on each time slot of each SS according to the channel state information, and obtains corresponding feature vectors and eigenvalues, and selects the largest eigenvalue and the corresponding feature vector to form each user end. The channel power gain matrix H of the time slot, the beamforming vector of the transmitting end corresponding to the matrix H, and the combining vector of the receiving end; the beamforming vector and the subcarrier allocation information are transmitted to the SS side through the BS control channel and the BS antenna; the power and bit allocation module is According to the beamforming vector and subcarrier allocation information transmitted by the BS, each SS is allocated a time slot by using a marginal utility and an improved greedy algorithm. The embodiment of the present invention adopts a method for adapting single-cell multi-user uplink adaptive sub-carrier, power and bit allocation, and guarantees the monthly quality requirements and fairness of each user end (SS, Subscriber Station). The carrier allocation is performed separately from the power and bit allocation, so that the base station (BS, Base Station) has the minimum transmission power, and at the same time, the transmission power of each SS is minimized, and the system performance is improved. The above is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or within the technical scope disclosed by the present invention. Alternatives are intended to be covered by the scope of the present invention.

Claims

Claim

A beamforming-based uplink system resource allocation method, the method comprising:

B. The BS allocates time slots for each SS according to the channel power gain matrix, and uses the marginal utility and the improved greedy algorithm; the BS sends the allocated time slots and the beamforming vectors corresponding to the time slots to the SSs;

D. Each SS sends the allocated power and bit information to the BS through the control part in the uplink subframe; the BS processes the received data to obtain the original data sent by the SS;

Repeat steps A through Steps for each subcarrier, power, and bit allocation allocation cycle

D.

The beamforming-based uplink system resource allocation method according to claim 1, wherein the step A specifically includes:

Al, the SS sends the service flow parameter to the BS when the service is customized;

A2. In each allocation period, the BS acquires channel state information of each time slot between the BS and each SS through channel sounding technology, and then performs eigenvalue decomposition on the MIMO channel matrix of each time slot of each SS to obtain a corresponding The feature vector and the eigenvalue are selected, and the largest eigenvalue and the corresponding eigenvector are selected to form a channel power gain matrix H of each SS slot, a transmitter beamforming vector corresponding to the matrix, and a merging vector.

3. The beamforming-based uplink system resource allocation method according to claim 1, wherein the time slot refers to:

Adopt adaptive modulation and coding in the uplink AMC subcarrier _ 方式 column mode, each time The slot is composed of 6 block bins and 1 orthogonal frequency division multiple access (OFDMA) symbol, and 1 bin is composed of 9 subcarriers in the OFDMA symbol, 8 of which are data subcarriers and the other is a pilot. Carrier. The beamforming-based uplink system resource allocation method according to claim 1, wherein the step B specifically includes:

B1. When each spatial subchannel on each time slot is transmitted in a maximum modulation order, the minimum number of slots required for each SS is calculated;

B2, first allocate a time slot for each SS;

B3. Find the top ranked time slot in each SS unassigned time slot, and calculate the marginal utility of the time slot;

B5. After determining step B4, whether there is a time slot remaining, if any, allocate the remaining time slot to the corresponding SS, otherwise, execute step «C. The beamforming-based uplink system resource allocation method according to claim 1, wherein the step C specifically includes:

Cl, SS uses the improved greedy algorithm for power and bit allocation on the assigned time slots;

C2, SS uses the allocation result in C1 to perform corresponding modulation and power adjustment, and weights the modulated symbols according to the corresponding beamforming vector, and then passes the inverse fast Fourier transform IFFT, parallel transform and prefix addition. BS. The beamforming-based uplink system resource allocation method according to claim 1, wherein the step D specifically includes:

D1, the receiving end first performs serial-to-parallel transformation and removes the cyclic prefix on the received data, and then performs fast Fourier transform FFT transform into frequency domain symbols, and the receiving end performs combined vector processing on the transformed frequency domain symbols;

D2. Demodulate the original data by using the received power and bit allocation information and the time slot allocation information. An apparatus for allocating uplink system resources based on beamforming, characterized in that the apparatus comprises an adaptive beamforming and subcarrier allocation module and a power and bit allocation module, wherein the adaptive beamforming and subcarrier allocation module is based on channel state information Performing eigenvalue decomposition on the MIMO channel matrix on each time slot of each SS to obtain a corresponding feature vector and eigenvalue, and selecting the largest eigenvalue and the corresponding eigenvector to form a channel power gain matrix of each SS slot. a matrix beam shaping vector corresponding to the matrix H and a receiving end combining vector; transmitting the beamforming vector and the subcarrier allocation information to the SS J through the BS control channel and the BS antenna;

The power and bit allocation module allocates time slots for each SS using marginal utility and improved greedy algorithm based on the beamforming vector and subcarrier allocation information transmitted by the BS.